LBNL Report Number
Genentech has ambitious energy and indoor environmental quality performance goals for Building 35 (B35) being constructed by Webcor at the South San Francisco campus. Genentech and Webcor contracted with the Lawrence Berkeley National Laboratory (LBNL) to test building systems including lighting, lighting controls, shade fabric, and automated shading controls in LBNL's new FLEXLAB facility. The goal of the testing is to ensure that the systems installed in the new office building will function in a way that reduces energy consumption and provides a comfortable work environment for employees.
LBNL tested three facades of the new office building in the rotating FLEXLAB testbed: west, south and east. External shading, lighting, and internal shading control was configured for each orientation to replicate the conditions of B35. The three facades were each tested for one week three times between July and October 2014. Changes were made between each test to improve the performance of the systems.
Linear pendant LED light fixtures will illuminate the open office areas of the office building. These fixtures were installed in FLEXLAB. The wide spacing between rows of light fixtures results in a low lighting power density of 0.57 W/ft2 in the open office areas, while still meeting the average illuminance criteria of 300 lux (28 footcandles). A combination of the wide spacing and optics of the light fixture creates a nonuniform lighting pattern on the ceiling of the space. Changing to a diffuse lens on the uplight will help reduce abrupt changes in luminance on the ceiling but non-uniformity will persist due to the wide spacing.
The pendant light fixtures allow separate control of the downward and upward light. The lighting control design aims to enhance the quality of space by dimming upward light unison providing uniform patterns of electric light on the ceiling. The downward light of each fixture dims to provide just enough light to meet illuminance criteria below the fixture.
Webcor installed two lighting control systems manufactured by Enlighted and Encelium for testing in FLEXLAB. The Encelium system uses an open loop control architecture with a ceiling-mounted photosensor at each facade (inside of the automated shade). While there is greater variation in workplane illuminance provided by the Encelium system, the system is better able to control upward versus downward lighting and is able to control the lighting according to the lighting design intent. The architecture of the Encelium system offers more functional flexibility by allowing any input (sensors, switches etc.) or multiple inputs to affect any fixture.
The Enlighted control system uses closed loop architecture with two photosensors per fixture (one for upward light and one for downward light). The Enlighted system controlled the lights more precisely than the Encelium system to meet workplane illuminance requirements, however the upward versus downward light control did not behave according to the lighting design intent.
MechoSystems provided motorized window shades and automated control. The shades in each window had a different color fabric, one dark grey and one medium grey. Both shade fabrics were an open weave with 3% openness. Genentech selected the dark colored shade because it provides a better view of the exterior compared to the lighter colored shade. Anecdotal evidence suggests that some occupants may experience direct glare with 3% open fabric while other occupants will not experience glare under the same conditions. Visual discomfort during the worst case sunny winter condition was not evaluated. However, the east-facing orientation during the equinox period was exposed to low sun angles in the third test period so findings of just acceptable visual discomfort are expected to be similar to what might be experienced during the winter.
The shades operated as expected on sunny days (which was the predominant condition during the test period). The testing identified substantial potential energy savings for the lighting systems by stopping the shade above the sill, preventing the shade from completely covering the window and allowing the sun to shine deeper into the space through the bottom few inches of the window. On partly cloudy days, which occurred more frequently after our testing concluded, anecdotal evidence suggests that the shades could be raised more often. LBNL suggests that a second threshold be implemented which drops the shade partway to prevent direct glare from bright sun, but doesn't close the shade down to the height required to limit sunshine depth.
Thermal comfort analysis suggests that occupants seated near the shaded window will be comfortable around 80% of the time. The 20% of time where the observed conditions fall outside the ASHRAE Standard 55 are almost always due to occupants being cold in the morning. This discomfort is mostly driven by cold surrounding surfaces causing a low mean radiant temperature and overcooling from outside air during economizer mode. Only one thermal comfort station, located near the facade, was used for the experiment. Thermal comfort further from the facade is unknown but is likely to be better due to the increased distance from the relatively cold facade.
Visual comfort studies indicated that occupants could sit as close as 3.5 feet to the east and west facade and 2.5 feet to the south facade when facing parallel to the window. Occupants must sit further away from the window to be comfortable when facing the window directly. Occupants should be 3.5 feet away when facing the south facade, 4.5 feet away when facing the west facade and 5.5 feet away when facing the east facade. Thermal comfort studies show that sitting within 30 inches of the facade has a negligible effect on comfort ratings.
Daylighting controls reduced lighting energy use in FLEXLAB by 46% for east facade, 34% for south facade and 35% for west facade over 30 feet deep perimeter zone between 7 AM and 7 PM local time at autumn equinox. Occupancy controls will further reduce lighting energy use, though they were not implemented for the test due to the cell being tested unoccupied.
Genentech, Webcor, and the architectural and engineering team had access to the FLEXLAB during and for a month following the test period to observe, work, and discuss operational issues with employees and staff. The project team made their own qualitative observations about the space in terms of view, adequacy of lighting and daylight levels, color, furniture placement, etc. The project team worked collaboratively with the LBNL team to fine tune details of component design, control settings, troubleshooting, and operations. Because Genentech is introducing a new model for their work environment, a non-assigned workplace, there were detailed discussions on how to educate the occupants about the new technologies and their operational modes. Commissioning and tuning procedures were also discussed.